Classification and Prediction (baseado nos slides do livro: Data Mining: C & T)

Classification and Prediction What is classification? What is prediction? Issues regarding classification and prediction Classification by decision tree induction Bayesian Classification (TI) Classification by Back Propagation (Neural Networks) (TI) Support Vector Machines (SVM) Associative Classification: Classification by association rule analysis Other Classification Methods: K-Nearest Neighbor, Case-based reasoning, etc Prediction: Regression (TI) 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification vs. Prediction predicts categorical class labels (discrete or nominal) classifies data (constructs a model) based on the training set and the values (class labels) in a classifying attribute and uses it in classifying new data Prediction models continuous-valued functions, i.e., predicts unknown or missing values Typical applications Credit approval Target marketing Medical diagnosis Fraud detection 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification—A Two-Step Process (1) Model construction: describing a set of predetermined classes Each tuple/sample is assumed to belong to a predefined class, as determined by the class label attribute The set of tuples used for model construction is the training set The model is represented as classification rules, decision trees, or mathematical formulae 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification—A Two-Step Process (2) Model usage: for classifying future or unknown objects Estimate accuracy of the model The known label of test sample is compared with the classified result from the model Accuracy rate is the percentage of test set samples that are correctly classified by the model Test set is independent of training set, otherwise over-fitting will occur If the accuracy is acceptable, use the model to classify data tuples whose class labels are not known 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification Process (1): Model Construction Algorithms Training Data Classifier (Model) IF rank = ‘professor’ OR years > 6 THEN tenured = ‘yes’ 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification Process (2): Use the Model in Prediction Classifier Testing Data Unseen Data (Jeff, Professor, 4) Tenured? 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Supervised vs. Unsupervised Learning Supervised learning (classification) The training data (observations, measurements, etc.) are accompanied by labels indicating the class of the observations New data is classified based on the training set Unsupervised learning (clustering) The class labels of training data is unknown Given a set of measurements, observations, etc. with the aim of establishing the existence of classes or clusters in the data 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification and Prediction What is classification? What is prediction? Issues regarding classification and prediction Classification by decision tree induction Bayesian Classification Classification by Back Propagation Support Vector Machines Associative Classification: Classification by association rule analysis 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Sistemas de Apoio à Decisão Data Preparation Data cleaning Preprocess data in order to reduce noise and handle missing values Relevance analysis (feature selection) Remove the irrelevant or redundant attributes Data transformation Generalize and/or normalize data 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Evaluating classification methods Accuracy: classifier accuracy and predictor accuracy Speed and scalability time to construct the model (training time) time to use the model (classification/prediction time) Robustness handling noise and missing values Scalability efficiency in disk-resident databases Interpretability understanding and insight provided by the model Other measures e.g., goodness of rules, such as decision tree size or compactness of classification rules Scalability evaluated by assessing the nb of I/O operations involved for a given classification algo. on data sets of increasingly large size. 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification Accuracy: Estimating Error Rates Partition: Training-and-testing use two independent data sets, e.g., training set (2/3), test set (1/3) used for data set with large number of samples Cross-validation divide the data set into k sub-samples use k-1 sub-samples as training data and one sub-sample as test data—k-fold cross-validation for data set with moderate size Bootstrapping (leave-one-out) for small size data 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Increasing Classfier Accuracy: Bagging General idea: averaging the prediction over a collection of classifiers Training data Altered Training data …….. Aggregation …. Classification method (CM) Classifier C CM Classifier C1 CM Classifier C2 Classifier C* 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Bagging: The Algorithm Given a set S of s samples Generate a bootstrap sample T from S. Cases in S may not appear in T or may appear more than once. Repeat this sampling procedure, getting a sequence of k independent training sets A corresponding sequence of classifiers C1, C2, …, Ck is constructed for each of these training sets, by using the same classification algorithm To classify an unknown sample X, let each classifier predict or vote The Bagged Classifier C* counts the votes and assigns X to the class with the “most” votes 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification and Prediction What is classification? What is prediction? Issues regarding classification and prediction Classification by decision tree induction Bayesian Classification (TI) Classification by Back Propagation (Neural Networks) (TI) Support Vector Machines (SVM) Associative Classification: Classification by association rule analysis Other Classification Methods: K-Nearest Neighbor, Case-based reasoning, etc Prediction: Regression (TI) 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Decision Tree Induction: Training Dataset This follows an example of Quinlan’s ID3 (Playing Tennis) Aprendizagem de um conceito: inferir o valor de uma função do cjto de treino, a partir do seu Input e output Função alvo: buys_computer Tarefa: prever o valor de buys_computer baseado nos valores dos outros atributos Representação de hipóteses: Hipótese – conjunção de restrições sobre os atributos Se uma instância satisfaz as restrições de uma hipóteses e buys_computer(x)) = 1 é ex. Positivo, Se buys_computer(x) = 0 então é exemplo negativo Aprendizagem é a tarefa de procurar num espaço de hipóteses aquela que melhor se adequa aos exemplos de treino 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Sistemas de Apoio à Decisão Output: A Decision Tree for “buys_computer” age? overcast student? credit rating? no yes fair excellent <=30 >40 30..40 Representa um conceito (comprar computador) ou seja, prevê se um cliente é um potencial comprador de computador. Árvore de decisão: representa uma disjunção de conjunções Para classificar uma amostra desconhecida, os valores dos seus atributos são testados contra a árvore de decisão. O caminho desde a raiz até ao nó folha é traçado e o nó folha dá a previsão de classe para essa amostra. Àrvores de decisão são adequadas para problemas onde: - as instâncias são descritas por pares atributo-valor; a função alvo tem valores discretos (existe extensão para valores contínuos); são necessárias descrições com disjunção; os dados de treino podem conter erros e pode conter missing values 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Algorithm for Decision Tree Induction (ID3) Basic algorithm (a greedy algorithm) Tree is constructed in a top-down recursive divide-and-conquer manner At start, all the training examples are at the root Attributes are categorical (if continuous-valued, they are discretized in advance) Examples are partitioned recursively based on selected attributes Test attributes are selected on the basis of a heuristic or statistical measure (e.g., information gain) Conditions for stopping partitioning All samples for a given node belong to the same class There are no remaining attributes for further partitioning – majority voting is employed for classifying the leaf There are no samples left A ideia de qq. Algo de indução de árvores de decisão é encontrar a árvore que melhor se adequa ao cjto de treino, isto é que melhor classifica o cjto de treino, dentro de todas as árvores possíveis. ID3 favorece árvores mis pequenas e aquelas que colocam os atributos com maior ganho de informação Mais perto da raiz. 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Sistemas de Apoio à Decisão Attribute Selection Measure: Information Gain (ID3/C4.5) Select the attribute with the highest information gain S contains si tuples of class Ci for i = {1, …, m} information measures info required to classify any arbitrary tuple entropy of attribute A with values {a1,a2,…,av} information gained by branching on attribute A Then how can we decide a test attribute on each node? One of the popular methods is using information gain measure, which we covered in chapter 5. It involves rather complicated equations, and I’ll not present the details here. Just basic ideas. The basic idea is that we select the attribute with the highest information gain. This information gain can be calculated from the expected information I and entropy of each attribute, E I : the expected information needed to classify a given sample E (entropy) : expected information based on the partitioning into subsets by A 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Attribute Selection by Information Gain Computation means “age <=30” has 5 out of 14 samples, with 2 yes’es and 3 no’s. Hence Similarly: Class P: buys_computer = “yes” Class N: buys_computer = “no” I(p, n) = I(9, 5) =0.940 Compute the entropy for age: 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Extracting Classification Rules from Trees Represent the knowledge in the form of IF-THEN rules One rule is created for each path from the root to a leaf Each attribute-value pair along a path forms a conjunction The leaf node holds the class prediction Rules are easier for humans to understand Example: IF age = “<=30” AND student = “no” THEN buys_computer = “no” IF age = “<=30” AND student = “yes” THEN buys_computer = “yes” IF age = “31…40” THEN buys_computer = “yes” IF age = “>40” AND credit_rating = “excellent” THEN buys_computer = “no” IF age = “<=30” AND credit_rating = “fair” THEN buys_computer = “yes” 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Avoid Overfitting in Classification Overfitting: An induced tree may overfit the training data Too many branches, some may reflect anomalies due to noise or outliers Poor accuracy for unseen samples Two approaches to avoid overfitting Prepruning: Halt tree construction early, do not split a node if this would result in the goodness measure falling below a threshold; difficult to choose an appropriate threshold Postpruning: Remove branches from a “fully grown” tree—get a sequence of progressively pruned trees Use a set of data different from the training data to decide which is the “best pruned tree” Overfitting pode acontecer se o cjto de treino é pequeno de mais Pode existir uma hipótese (árvore) que se comporte melhor com os dados completos, Embora pior com os dados de treino. 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Approaches to Determine the Final Tree Size Separate training (2/3) and testing (1/3) sets Use cross validation, e.g., 10-fold cross validation Use all the data for training but apply a statistical test (e.g., chi-square) to estimate whether expanding or pruning a node may improve the entire distribution Use minimum description length (MDL) principle halting growth of the tree when the encoding is minimized 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Enhancements to Basic Decision Tree Induction Allow for continuous-valued attributes Dynamically define new discrete-valued attributes that partition the continuous attribute value into a discrete set of intervals Handle missing attribute values Assign the most common value of the attribute Assign probability to each of the possible values Attribute construction Create new attributes based on existing ones that are sparsely represented This reduces fragmentation, repetition, and replication 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Computing Info. Gain for Continuous-Value Attributes Let attribute A be a continuous-valued attribute Must determine the best split point for A Sort the value A in increasing order Typically, the midpoint between each pair of adjacent values is considered as a possible split point (ai+ai+1)/2 is the midpoint between the values of ai and ai+1 The point with the minimum expected information requirement for A is selected as the split-point for A Split: D1 is the set of tuples in D satisfying A ≤ split-point, and D2 is the set of tuples in D satisfying A > split-point 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Classification in Large Databases Classification: a classical problem extensively studied by statisticians and machine learning researchers Scalability: Classify data sets with millions of examples and hundreds of attributes with reasonable speed Why decision tree induction in data mining? relatively faster learning speed (than other classification methods) convertible to simple and easy to understand classification rules can use SQL queries for accessing databases comparable classification accuracy with other methods 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Scalable Decision Tree Induction Methods SLIQ (EDBT’96 — Mehta et al.): builds an index for each attribute and only class list and the current attribute list reside in memory SPRINT (VLDB’96 — J. Shafer et al.): constructs an attribute list data structure PUBLIC (VLDB’98 — Rastogi & Shim): integrates tree splitting and tree pruning: stop growing the tree earlier RainForest (VLDB’98 — Gehrke, Ramakrishnan & Ganti): separates the scalability aspects from the criteria that determine the quality of the tree; builds an AVC-list (attribute, value, class label) 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Presentation of Classification Results 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Visualization of a Decision Tree in SGI/MineSet 3.0 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Sistemas de Apoio à Decisão Bibliografia (Livro) Data Mining: Concepts and Techniques, J. Han & M. Kamber, Morgan Kaufmann, 2001 (Secções 7.1 a 7.3 – livro 2001, Secções 5.1 a 5.3 – draft) (Livro) Machine Learning, T. Mitchell, McGraw Hill, 1997 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)

Data Cube-Based Decision-Tree Induction Integration of generalization with decision-tree induction (Kamber et al’97). Classification at primitive concept levels E.g., precise temperature, humidity, outlook, etc. Low-level concepts, scattered classes, bushy classification-trees Semantic interpretation problems. Cube-based multi-level classification Relevance analysis at multi-levels. Information-gain analysis with dimension + level. 2003/04 Sistemas de Apoio à Decisão (LEIC Tagus)